Integrated value chain risk-based profiling and optimization

Abstract

A system and method for gathering and analyzing the value chain relationships between legal entities, people, systems, and real and intangible assets using a temporospatial knowledge graph of the integrated value chain. The system provides the ability to layer private data from paid vendors with end-user owned and public records data to enable more comprehensive, contextualized and complete representations of the underlying value chain. Data analysis techniques, such as deep learning and machine learning, are performed on the knowledge graph and its underlying data set, in conjunction with simulation and modeling, to analyze the value chain, including generation of a risk profile for an entity's value chain and potential optimization options to remediate the identified risks.

What is claimed is: 
     
         1 . A system for integrated value chain risk-based profiling and optimization, comprising:
 a computing device comprising a memory, a processor, and a non-volatile data storage device;   an integrated value chain ontological database comprising value chain data; and   a directed computational graph module comprising a first plurality of programming instructions stored in a memory of, and operating on a processor of, a computing device, wherein the first plurality of programming instructions, when operating on the processor, cause the computing device to:
 analyze the integrated value chain ontological database for risk query related information, the query-related information comprising entities, individuals, locations, and topics associated with the subject; and 
 create a weighted and directed temporospatial knowledge graph, the weighted and directed temporospatial knowledge graph comprising nodes representing the entities, individuals, locations, and topics associated with the subject and edges representing the relationships to the nodes in relation to the subject or the associated nodes; and 
   a risk optimization engine comprising a second plurality of programming instructions stored in a memory of, and operating on a processor of, a computing device, wherein the second plurality of programming instructions, when operating on the processor, cause the computing device to:
 identify paths or clusters of a subset of the weighted and directed temporospatial knowledge graph which meet a pre-determined risk threshold wherein the paths or clusters represent risk categories; 
 perform one or more simulations using data from at least part of the integrated value chain ontological database wherein:
 the simulations model a disruption event to determine a probability and disruption impact associated with the disruption event; 
 the simulation models alternative actions to determine the feasibility of the alternative action; and 
 
   assign a risk score to each identified risk category, based on the probability and disruption impact associated with that risk.   
     
     
         2 . The system of  claim 1 , further comprising an asset registry manager comprising a third plurality of programming instructions stored in a memory of, and operating on the processor of, the computing device, wherein the third plurality of programming instructions, when operating on the processor, cause the computing device to:
 scan ingested data from information sources for provenance metadata;   store the provenance metadata on the non-volatile storage device; and   send the provenance metadata to an automated ontology engine.   
     
     
         3 . The system of  claim 1 , further comprising an automated ontology engine comprising a fourth plurality of programming instructions stored in a memory of, and operating on the processor of, the computing device, wherein the fourth plurality of programming instructions, when operating on the processor, cause the computing device to:
 receive data collected using web-scrapers and data-extraction tools, from all available data sources; and   receive the provenance metadata from the asset registry manager;   run the collected data through a variety of tools to append the data with temporal, geospatial, information reliability, contextual metadata, and provenance metadata using machine learning algorithms and ontological axioms configuration; and   store the appended data in the integrated value chain ontological database.   
     
     
         4 . The system of  claim 3 , wherein the data sources include at least parts of public, private, and proprietary data sources. 
     
     
         5 . The system of  claim 1 , wherein the simulation is a Monte Carlo simulation. 
     
     
         6 . The system of  claim 1 , wherein the simulation data further comprises partial synthetic data. 
     
     
         7 . The system of  claim 1 , where the simulations model a fraud event, and assigns a fraud risk score to a fraud category. 
     
     
         8 . A method for integrated value chain risk-based profiling and optimization, comprising the steps of:
 analyzing the integrated value chain ontological database for risk query related information, the query-related information comprising entities, individuals, locations, and topics associated with the subject;   creating a weighted and directed temporospatial knowledge graph, the weighted and directed temporospatial knowledge graph comprising nodes representing the entities, individuals, locations, and topics associated with the subject and edges representing the relationships to the nodes in relation to the subject or the associated nodes;   identifying paths or clusters of a subset of the weighted and directed temporospatial knowledge graph which meet a pre-determined risk threshold wherein the paths or clusters represent risk categories;   performing one or more simulations using data from at least part of the integrated value chain ontological database wherein:
 the simulations model a disruption event to determine a probability and disruption impact associated with the disruption event; 
 the simulation models alternative actions to determine the feasibility of the alternative action; and 
 assigning a risk score to each identified risk category, based on the probability and disruption impact associated with that risk. 
   
     
     
         9 . The method of  claim 8 , performing the following steps using an asset registry manager, the asset registry manager comprising a first memory, a first processor, and a third plurality of programming instructions:
 scanning ingested data from information sources for provenance metadata;   storing the provenance metadata on the non-volatile storage device; and   sending the provenance metadata to an automated ontology engine.   
     
     
         10 . The method of  claim 8 , performing the following steps using an automated ontology engine, the automated ontology engine comprising a second memory, a second processor, and a fourth plurality of programming instructions:
 receiving data collected using web-scrapers and data-extraction tools, from all available data sources; and   receiving the provenance metadata from the asset registry manager;   running the collected data through a variety of tools to append the data with temporal, geospatial, information reliability, contextual metadata, and provenance metadata using machine learning algorithms and ontological axioms configuration; and   storing the appended data in the integrated value chain ontological database.   
     
     
         11 . The method of  claim 8 , wherein the data sources include at least parts of public, private, and proprietary data sources. 
     
     
         12 . The method of  claim 8 , wherein the simulation is a Monte Carlo simulation. 
     
     
         13 . The method of  claim 8 , wherein the simulation data further comprises partial synthetic data. 
     
     
         14 . The method of  claim 8 , where the simulations model a fraud event, and assigns a fraud risk score to a fraud category.